Narrative:

After an uneventful landing and clearing runway at ZZZ we received a right pack high temperature EICAS caution. The packs were being operated in automatic mode and the caution went away just seconds after when, as I learned after the event, first officer had reset the right pack. The same message came back roughly 15 seconds later and I immediately verified it by selecting the environmental control system synoptic page on ED2. As I did so flight attendant called the cockpit via the emergency call function and the first officer answered the call as I brought the aircraft to a stop on the taxiway. By the time I could shift my focus from stopping the aircraft to selecting the proper comm switches, the flight attendant had nearly completed her description of 'smoke in the cabin.' the exact instant that I heard her say 'smoke' I smelled it as well. It smelled like rubber or metal and my first thought was of an electrical fire. I understood that it was coming from the back of the aircraft and that knowledge, combined with the fact that I smelled it up front just seconds after it was first discovered, led me to believe that it was filling the cabin at a moderate rate. I also understood that it was definitely visible yet not overpowering or blinding. Based on this assessment I decided to execute a rapid 'normal' deplaning using the passenger door only. I asked first officer to notify the tower of the smoke in the cabin, ask for the airport fire equipment, and inform them that the passengers would deplane immediately on the taxiway. After completing this he started into the emergency evacuation checklist a few seconds prior to my command. For a near brilliant, valid reasons, which I will detail toward the end of the report, I think he was anxious to start the checklist while I was growing somewhat confused as to the amount of smoke in the cabin. Keep in mind that the events to this point in this paragraph took only 5-10 seconds. At the time I thought it was more important to make the proper assessment of the situation than to make a costly error and save 10 seconds on the completion of the checklist. In retrospect, I back this decision. As flight attendant performed her duties, I caught up to first officer on the emergency evacuation checklist. Because of my prior confusion on the exact status of the smoke, when we got to the point of the 'evacuate' verbal PA on the checklist, I decided I better perfect the communication between myself and the flight attendant so I opened the cockpit door. I felt it would be tragic if we needed more exits and did not use them. I asked the flight attendant if we could get everyone out of the passenger door safely or if we needed to utilize the over-wing emergency exits. She said that the passenger door was sufficient and at this point I could see that the smoke was not so thick as to incapacitate. I was also able to note that flight attendant had the situation well under control so I made no PA call to the cabin, I continued with the checklist and the aircraft was secure. The smoke seemed to stop entering the cabin, most likely when the power was fully removed from the aircraft. The passengers all complied with crew instructions and were nicely and safely contained on the grass strip between taxiways. In retrospect I am thrilled and even rather amazed at how well we kept them together in a group. Post event, the flight attendant told me that the passengers were the first to notice the smoke from the back and they shouted out to alert her. Everyone was accounted for and taken to the terminal by a mobile lounge. The fire and operations trucks arrived so quickly that it seemed to be less than one min from the radio call. The lone mechanic to arrive to the aircraft while I was still there took a quick look around. He told us that the APU battery had a melt-down and that the smoke most likely entered the cabin since the outflow valves automatically open after landing. Flight attendant performed at the highest possible level including total command of her passengers and emphatic instruction on how she expected them to accomplish their assignments. In analysis of the event, I find myself asking just two questions. Why did I make the decision to perform an 'orderly deplaning' and then seconds later question whether that was going to be fast enough to get everyone off? However, the major reason for the imperfect communication is the poor performance of the cl-65 intercom. As is the case with every other rj in our fleet, I could barely hear her. Some have suggested that it's my headset, but I don't think so. I have the same telex airman that everyone else has. I think the cockpit speaker proves my theory when, in high cruise flight or on the ground at the gate, one must turn the communication radio volume way down and the speaker volume all the way up just the hear the flight attendant. The volume transmitted by the flight attendant's handset seems to be about half that of the communication radios and the intercom between the pilots at the same setting. I only have 110 hours in the aircraft, nevertheless I have yet to understand a flight attendant clearly without tuning out the communication radio and selecting the intercom volume to full. I think we need to have something written that separates the orderly deplaning from the evacuation. Maybe the procedure would be something that allows the engines to be sent shut down sooner or something with a different call to the cabin to signal to the flight attendant that it is now safe outside. Callback conversation with reporter revealed the following information: the reporter stated the battery was a nickel cadmium type and had melted down causing the smoke in the cabin. The reporter said the smoke entered the passenger cabin attendant from the open outflow valves located near the batteries. The reporter stated the valves were opened on landing ground sensing. The reporter said no EICAS primary display was received in the cockpit for the overheat condition or were any circuit breakers or current limiters opened or tripped. The reporter stated the evacuation went orderly and no injuries were reported.

Related reports:

Google
 

Original NASA ASRS Text

Title: A CANADAIR CL65 AFTER CLRING THE RWY THE CAB ATTENDANT RPTED SMOKE IN THE CAB. ACFT WAS EVACUATED ON THE TXWY. CAUSED BY APU BATTERY MELT DOWN.

Narrative: AFTER AN UNEVENTFUL LNDG AND CLRING RWY AT ZZZ WE RECEIVED A R PACK HIGH TEMP EICAS CAUTION. THE PACKS WERE BEING OPERATED IN AUTO MODE AND THE CAUTION WENT AWAY JUST SECONDS AFTER WHEN, AS I LEARNED AFTER THE EVENT, FO HAD RESET THE RIGHT PACK. THE SAME MSG CAME BACK ROUGHLY 15 SECONDS LATER AND I IMMEDIATELY VERIFIED IT BY SELECTING THE ENVIRONMENTAL CTL SYS SYNOPTIC PAGE ON ED2. AS I DID SO FA CALLED THE COCKPIT VIA THE EMER CALL FUNCTION AND THE FO ANSWERED THE CALL AS I BROUGHT THE ACFT TO A STOP ON THE TXWY. BY THE TIME I COULD SHIFT MY FOCUS FROM STOPPING THE ACFT TO SELECTING THE PROPER COMM SWITCHES, THE FA HAD NEARLY COMPLETED HER DESCRIPTION OF 'SMOKE IN THE CABIN.' THE EXACT INSTANT THAT I HEARD HER SAY 'SMOKE' I SMELLED IT AS WELL. IT SMELLED LIKE RUBBER OR METAL AND MY FIRST THOUGHT WAS OF AN ELECTRICAL FIRE. I UNDERSTOOD THAT IT WAS COMING FROM THE BACK OF THE ACFT AND THAT KNOWLEDGE, COMBINED WITH THE FACT THAT I SMELLED IT UP FRONT JUST SECONDS AFTER IT WAS FIRST DISCOVERED, LED ME TO BELIEVE THAT IT WAS FILLING THE CABIN AT A MODERATE RATE. I ALSO UNDERSTOOD THAT IT WAS DEFINITELY VISIBLE YET NOT OVERPOWERING OR BLINDING. BASED ON THIS ASSESSMENT I DECIDED TO EXECUTE A RAPID 'NORMAL' DEPLANING USING THE PAX DOOR ONLY. I ASKED FO TO NOTIFY THE TWR OF THE SMOKE IN THE CABIN, ASK FOR THE ARPT FIRE EQUIP, AND INFORM THEM THAT THE PASSENGERS WOULD DEPLANE IMMEDIATELY ON THE TXWY. AFTER COMPLETING THIS HE STARTED INTO THE EMER EVACUATION CHKLIST A FEW SECONDS PRIOR TO MY COMMAND. FOR A NEAR BRILLIANT, VALID REASONS, WHICH I WILL DETAIL TOWARD THE END OF THE RPT, I THINK HE WAS ANXIOUS TO START THE CHKLIST WHILE I WAS GROWING SOMEWHAT CONFUSED AS TO THE AMOUNT OF SMOKE IN THE CABIN. KEEP IN MIND THAT THE EVENTS TO THIS POINT IN THIS PARAGRAPH TOOK ONLY 5-10 SECONDS. AT THE TIME I THOUGHT IT WAS MORE IMPORTANT TO MAKE THE PROPER ASSESSMENT OF THE SIT THAN TO MAKE A COSTLY ERROR AND SAVE 10 SECONDS ON THE COMPLETION OF THE CHKLIST. IN RETROSPECT, I BACK THIS DECISION. AS FA PERFORMED HER DUTIES, I CAUGHT UP TO FO ON THE EMER EVACUATION CHKLIST. BECAUSE OF MY PRIOR CONFUSION ON THE EXACT STATUS OF THE SMOKE, WHEN WE GOT TO THE POINT OF THE 'EVACUATE' VERBAL PA ON THE CHKLIST, I DECIDED I BETTER PERFECT THE COM BTWN MYSELF AND THE FA SO I OPENED THE COCKPIT DOOR. I FELT IT WOULD BE TRAGIC IF WE NEEDED MORE EXITS AND DID NOT USE THEM. I ASKED THE FA IF WE COULD GET EVERYONE OUT OF THE PASSENGER DOOR SAFELY OR IF WE NEEDED TO UTILIZE THE OVER-WING EMER EXITS. SHE SAID THAT THE PAX DOOR WAS SUFFICIENT AND AT THIS POINT I COULD SEE THAT THE SMOKE WAS NOT SO THICK AS TO INCAPACITATE. I WAS ALSO ABLE TO NOTE THAT FA HAD THE SIT WELL UNDER CTL SO I MADE NO PA CALL TO THE CABIN, I CONTINUED WITH THE CHKLIST AND THE ACFT WAS SECURE. THE SMOKE SEEMED TO STOP ENTERING THE CABIN, MOST LIKELY WHEN THE POWER WAS FULLY REMOVED FROM THE ACFT. THE PASSENGERS ALL COMPLIED WITH CREW INSTRUCTIONS AND WERE NICELY AND SAFELY CONTAINED ON THE GRASS STRIP BTWN TAXIWAYS. IN RETROSPECT I AM THRILLED AND EVEN RATHER AMAZED AT HOW WELL WE KEPT THEM TOGETHER IN A GROUP. POST EVENT, THE FA TOLD ME THAT THE PASSENGERS WERE THE FIRST TO NOTICE THE SMOKE FROM THE BACK AND THEY SHOUTED OUT TO ALERT HER. EVERYONE WAS ACCOUNTED FOR AND TAKEN TO THE TERMINAL BY A MOBILE LOUNGE. THE FIRE AND OPS TRUCKS ARRIVED SO QUICKLY THAT IT SEEMED TO BE LESS THAN ONE MIN FROM THE RADIO CALL. THE LONE MECHANIC TO ARRIVE TO THE ACFT WHILE I WAS STILL THERE TOOK A QUICK LOOK AROUND. HE TOLD US THAT THE APU BATTERY HAD A MELT-DOWN AND THAT THE SMOKE MOST LIKELY ENTERED THE CABIN SINCE THE OUTFLOW VALVES AUTOMATICALLY OPEN AFTER LNDG. FA PERFORMED AT THE HIGHEST POSSIBLE LEVEL INCLUDING TOTAL COMMAND OF HER PASSENGERS AND EMPHATIC INSTRUCTION ON HOW SHE EXPECTED THEM TO ACCOMPLISH THEIR ASSIGNMENTS. IN ANALYSIS OF THE EVENT, I FIND MYSELF ASKING JUST TWO QUESTIONS. WHY DID I MAKE THE DECISION TO PERFORM AN 'ORDERLY DEPLANING' AND THEN SECONDS LATER QUESTION WHETHER THAT WAS GOING TO BE FAST ENOUGH TO GET EVERYONE OFF? HOWEVER, THE MAJOR REASON FOR THE IMPERFECT COM IS THE POOR PERFORMANCE OF THE CL-65 INTERCOM. AS IS THE CASE WITH EVERY OTHER RJ IN OUR FLEET, I COULD BARELY HEAR HER. SOME HAVE SUGGESTED THAT IT'S MY HEADSET, BUT I DON'T THINK SO. I HAVE THE SAME TELEX AIRMAN THAT EVERYONE ELSE HAS. I THINK THE COCKPIT SPEAKER PROVES MY THEORY WHEN, IN HIGH CRUISE FLT OR ON THE GND AT THE GATE, ONE MUST TURN THE COM RADIO VOLUME WAY DOWN AND THE SPEAKER VOLUME ALL THE WAY UP JUST THE HEAR THE FA. THE VOLUME TRANSMITTED BY THE FA'S HANDSET SEEMS TO BE ABOUT HALF THAT OF THE COM RADIOS AND THE INTERCOM BTWN THE PLTS AT THE SAME SETTING. I ONLY HAVE 110 HRS IN THE ACFT, NEVERTHELESS I HAVE YET TO UNDERSTAND A FA CLEARLY WITHOUT TUNING OUT THE COM RADIO AND SELECTING THE INTERCOM VOLUME TO FULL. I THINK WE NEED TO HAVE SOMETHING WRITTEN THAT SEPARATES THE ORDERLY DEPLANING FROM THE EVACUATION. MAYBE THE PROC WOULD BE SOMETHING THAT ALLOWS THE ENGS TO BE SENT SHUT DOWN SOONER OR SOMETHING WITH A DIFFERENT CALL TO THE CABIN TO SIGNAL TO THE FA THAT IT IS NOW SAFE OUTSIDE. CALLBACK CONVERSATION WITH RPTR REVEALED THE FOLLOWING INFO: THE RPTR STATED THE BATTERY WAS A NICKEL CADMIUM TYPE AND HAD MELTED DOWN CAUSING THE SMOKE IN THE CABIN. THE RPTR SAID THE SMOKE ENTERED THE PAX CAB FROM THE OPEN OUTFLOW VALVES LOCATED NEAR THE BATTERIES. THE RPTR STATED THE VALVES WERE OPENED ON LNDG GND SENSING. THE RPTR SAID NO EICAS PRIMARY DISPLAY WAS RECEIVED IN THE COCKPIT FOR THE OVERHEAT CONDITION OR WERE ANY CIRCUIT BREAKERS OR CURRENT LIMITERS OPENED OR TRIPPED. THE RPTR STATED THE EVACUATION WENT ORDERLY AND NO INJURIES WERE RPTED.

Data retrieved from NASA's ASRS site as of July 2007 and automatically converted to unabbreviated mixed upper/lowercase text. This report is for informational purposes with no guarantee of accuracy. See NASA's ASRS site for official report.